Real-Time Obstacle Detection in 3D Environments Using P300-based EEG Signals

Abstract

The research introduces a new system for obstacle detection in 3D spaces using P300 eventrelated potential from EEG signals. By incorporating brain-computer interfaces with robotics, this method enhances intuitive human-robot interaction. The system utilizes the Common Spatial Pattern (CSP) algorithm to identify distinct EEG features, improving the safety and adaptability of robots. Tested in a custom 3D simulation environment, the system demonstrated high accuracy and real-time performance, indicating its potential for applications like autonomous navigation and assistive robotics. This study represents a significant step in robotics and brain-computer interfaces, fostering a more natural user interface and propelling advancements in the field.